Chicago drowning in trains

[Anonymous]

Good insights O.S.

[Anonymous]

Tom: Incisive question. Many people thought that the gross ton-mile growth we're experiencing wouldn't happen, because heavy capital goods and durable consumer goods would achieve market saturation, then become more durable, nearly eliminating new installations and reducing replacement needs. That is, as soon as everyone bought a washing machine, we wouldn't be shipping so much weight across the country -- just a few replacement machines as the old ones wore out.

That was pretty naive thinking. In fact, heavy and durable goods did decline, once everyone had one of everything, but in the meantime, transportation capacity had been built to haul all that stuff, and now it had spare capacity. That allowed people to leverage that spare capacity and transportation productivity to source things from a lot farther away that we used to produce locally. So what we've quit shipping in terms of heavy goods we've filled right back up with things like food, clothing, paper goods, and other consumables. To leverage labor costs, the U.S. and Canada will continue to source things from farther and farther away, and transportation GTMs will continue to grow. In the very long run, wages in foreign countries will rise a little, while ours fall off a cliff to meet them at the bottom, and these labor advantages of foreign countries will evaporate. But we'll still have a huge transportation infrastructure in place which can move things very cheaply, so then natural geographic advantages will come into play.

While there might be a day in the future when averaged GTMs begin to decline, I don't think it will be in my lifetime. Wages are already on the decline, though.

OS

[Anonymous]

Wondering about the growth projections. Are people going to continue to load up their lives with so much physical stuff? Will the transportation component trend downward as technological change continues? Projection TV's become thin membranes?

[Anonymous]

Gary: That is essentially true. I think their inclusion of "aircraft" in the sentence shows they don't know the difference between freight value and freight tonnage. A huge percentage of freight value moves by air, but the effects of that freight on the highway and rail infrastructure is almost nil.

Greyhounds: I'm suprised to see you advance this argument. Your other posts show great awareness of complexity. You would propose that all the intermodal traffic reaching Chicago should simply be rubbered from one side to another, each truck paying a fuel tax that is penny or so on the dollar of its actual use of roadway resources, and that the air pollution and subsequent impact on everyone else's health care, building maintenance costs, etc., should be foisted onto the people paying for the health care and building maintenance?

Is there ANY government expenditure you think justified? National defense -- shouldn't that be paid for in user fees, too? I realize that your insistence on belittling those who disagree with you by calling it "gubermint" means you're probably not open to different opinions, but I'll give you the benefit of any doubt.

I agree, in the long run, the free market will work just fine. Traffic and pollution will drive people out of Chicago, real estate prices will collapse, and the railroads will find it easy to build connections across the deserted land once occupied by a thriving metropolis. The loss of wealth will be enormous, the loss of productivity will be enormous, but measured against the threat to a sacred ideology, a trillion dollars here or there seems a cheap price indeed. My friend, I don't like the abuse of the free market by people using the government to extract money from my wallet to put in their wallet any more than you do -- and I spend a lot of time in my job analyzing the negative (and sometimes positive) effects of subsidy, regulation, and taxation, but in my view, the whole point of life isn't to erect a shrine to the free market, it's to be happy, healthy, free of sin, and free of fear -- you know, the preface to the U.S. Constitution. The free market isn't the end we seek, it's a means to our end.

OS

[Anonymous]

Good explanations for why there’s a problem. I’ll throw in another quote from the article. Air-freight tonnage in Chicago is projected to double in the next 25 years; the number of lorries coming through the area is expected to rise by 80%; rail volume should almost double in two decades. And there will be a compounding effect from the fast growth of inter-modal transport (switching freight from one “mode” to another, such as train to truck or aircraft) which tends to send even more traffic to hubs like Chicago

[eolafan]

QUOTE: Originally posted by BNSF railfan.

Sounds like it's time to build a new by pass around the windy city.

You may find it interesting that the EJ&E is very much busier in the past two years than ever before, serving like the bypass you speak about. I heard a rumor last weekend about them looking at possibly double tracking "The J" in the future to accomodate more bypass traffic, but it was just a rumor. The UP and CN are using the J right now in addition to the J's own trains. I live about a mile from their main and hear lots of trains at night blowing for the Rt. 34 crossing. Also, when I cross Rt. 34 I see their approach lit signals glowing much more often than ever before.

PS: Chicago drowing in trains may be not so good for the railroads but very good for us railfans.

[greyhounds]

They've been trying to fix rail congestion in Chicago ever since the railroads got to Chicago. If you want to see trains standing still, this is the place. I ride in every workday on Metra. Lots of waiting freights.

A real hang up is Tower A-2 at Western Avenue. That's where the old Milwaukee Road commuter lines, plus the new North Central service, cross the old C&NW west commuter line.

East of the tower the tracks are seven wide, 3 for the MILW and 4 for the CNW.
Topping the whole thing off is that the storage/maintenance facility for the old CNW trains is west of the interlocking. So even trains that run other ex-CNW routes have to go through the interlocking before and after their inbound/outbound runs. There are a lot of stopped trains waiting at the junction as I ride through.

Right at tower A-2 there is an industry that actually still uses boxcars. It must be fun/expensive/interesting to get those cars in and out.

The comment about "Good luck getting money to fix this from the Bush Administration" is ignorance. For the Federal Government to fund this they would have to take the money from other economic development in the private sector. That is, they'd have to tax it away from some other purpose.

The problem is, they have absolutely no, none, nada, zero way of knowing what they are taking it away from. And they can't begin to measure the cost/benifits of doing such a thing. That's why the gubernmint should stay The Hell! out of allocating economic resources.

Absent gubernmint intrustion, investment capital is basically allocated on an auction basis. That lets each applicant for capital decide how much the allocation is worth to him/her. It isn't perfect, nothing is, but it sure beats a bunch of politicians deciding which projects get funded.

If an improvement in Chicago, or anywhere else, is worthwhile it will be privately financed. Otherwise, it just is less costly to let the trains sit.

[ericsp]

If one has to drown, drowning in trains is not a bad way to go.

[Anonymous]

Short and sweet, Mr. Husman.

OS

[dehusman]

Garyaiki said:
"Before a container ship arrives in port a plan is drawn up to unload, transfer, and reload the ship in the shortest possible time. It looks like no one is responsible for this when trains arrive in Chicago. "
Every railroad has a plan for every train. When it gets to a yard there is a classification system in place. Every destination is served by a different train. If a train arrives with a car for the ABC RR, then the car will leave on train 102. If a train arrives with a ca for the ABC RR, then the car will depart on train 102 tomorrow too. There is a plan in place.
Part of the problem is sheer volume. The other is that with a container ship the cans drive off by themselves. Railroad cars don't do that.
The container ship analogy breaks down because its not the same operation. A steamship docking is INTERmodal, the railway interchange in Chicago is INTRAmodal. That makes a big difference in the dynamics.

Dave H.

[Anonymous]

Ever wonder how a B747 can fall out of the sky if only one part breaks? It's because all the other parts can't carry on without it. Railroads are no different. One thing breaks, and the whole system starts to grind to a halt. The railroad version of flight is the fixed guideway: you can't steer around it and continue on as if nothing happened, and you can't do without it. Plus, the locomotives, cars, and crew are all already being figured for future trains, so you're affecting future trains, not just the ones you have on hand. Railroads run 24 hours a day -- they have no makeup time every night like an airline or your average workplace, where the timer can be reset to zero for the next morning and you start over and have chronological forgiveness for everything that went out of kilter the day before.

And when something goes wrong on a railroad, it isn't right there next to you in the office where you can reach out and grab it. Or out on the factory floor under lights, temperature control, and clean conditions. It's 423 miles away from you, 20 minutes on foot from the nearest 4WD road, in the middle of the night, in the mud, in a pouring rainstorm, in a windstorm, the power has failed, the microwave carrier for the radio has failed, and your nearest maintainer has 92 minutes left on his hours of service, it's a Saturday morning, the roadmaster is on vacation, the signal maintainer is laid up with a broken ankle, and you've got 18 trains holding while you're wondering what the hell to do next as one of them reports 24 inches of water over the rail and another reports a bank failure along a cresting river. Standing still starts to seem like a bad idea, but do you dare move one of them ... right into a washout?

Oh -- I'm not dreaming this up. It describes one night a few weeks ago. It wasn't all that unusual a night on anyone's railroad.

Ever throw a rock into a small, still puddle, and see the ripples reflect and refract for the next minute? Write a mathematical formula that predicts the magnitude, frequency, and position of each wave crest one minute into the future, accounting for all the variations in the shoreline and water depth of that pond. Essentially it describes how one car can affect one railroad on one day. Now drop 100,000 pieces of gravel into a good sized pond all at once, and write a mathematical formula that predicts the surface of the pond 1 second, 5 seconds, 10 seconds, and one minute later. That describes the typical day on a Class I railroad. We know in broad terms what's going to happen -- a bunch of ripples. In specific terms, no one has a clue where each wave crest will be, when.

Mathematically speaking, railroads are nonlinear dynamic systems that give rise to chaotic and unpredictable results that oscillate within broad boundaries. Every single variable has influence on every other variable because of the nature of the fixed guideway. Consider these unpredictabilities that are all in play for one train on one mile of average track at any given instant:

1. Discontinuity of track -- broken rail, washout, track buckle, wide gauge, etc.
2. Foreign object on track -- boulder, large animal, motor vehicle, trespasser
3. Failure of locomotives -- each a system of many thousand dependent parts
4. Failure of any one of 100+ cars -- each a system of a few hundred dependent parts
5. Failure of signal system -- itself reliant on long-distance communication systems
6. Failure of line power -- no signal power results
7. Failure of human crew -- they fall asleep, don't show up, miss a signal, etc.
8. Failure of dispatcher -- he focuses on something else, gets behind, etc.
9. Failure of fuel -- it happens!

And so forth: over 100,000 interdependent parts all present in a single train and a single mile of track of which the failure of any one has the substantial if not absolute potential to stop that train, and thus affect every other train to a distance of hundreds of miles away. If the train is stopped, for say, just one hour, the crew, for instance, now ties up one hour later, which means your ability to use them again 12 hours later has been altered. It affects every train it connects with. It affects every train it meets with. It may result in some of the cars missing connections, dwelling in a yard 24 hours, occupying a track the YM would usually use for something else, spilling into effects on how that yard works, delaying five more trains a few minutes each, etc.

This page describes some of the efforts to describe non-linear dynamic systems with math: http://en.wikipedia.org/wiki/Logistic_map Don't ask me to explain it; this level of math is about as far as I can grasp in a very general sense.

If it was as simple as an IT problem or a central dispatching office, the railroads would have done it already. What would be the point of staffing a facility to consider information that can't be predicted? Some insufferable twit from the Economist shows up, sees a fax machine instead of some fancy workstations, and tut-tuts at the ignorance of these poor backward railroaders. Perhaps his self-importance blinded him to the possibility that railroads do it this way because anything more elaborate is a waste of money that would create more problems than it solved. But, having already filled his head with all the knowledge he'll ever need, he simply compares railroads against airlines, like railroads are some sort of brutish stone-age tribe and airlines the "civilized world." I think I'd like to ask him, "If airlines are so clever and sophisticated, how is it that their net profit since Wilbur Wright fluttered into the air is $0.00?

OS

[Anonymous]

Thanks OS for a clear explanation of the physical constraints of routing through Chicago but the thing about the article that’s bothered me for a week was this sentence. Unlike the airlines, railroads don't have an oversight agency to co-ordinate their traffic: each of them uses its own dispatchers, who sometimes send faxes to their counterparts to let them know trains are coming. That’s an IT problem not a physical problem. Before a container ship arrives in port a plan is drawn up to unload, transfer, and reload the ship in the shortest possible time. It looks like no one is responsible for this when trains arrive in Chicago. If that’s the case, inter-railroad information technology would expedite cargo through Chicago, to some extent, without building bypasses or running short-tonnage trains.

[BaltACD]

OS - well presented

I might add that the Class I carriers have and and continue to implement system wide blocking schemes that utilize Gateways other than Chicago to all extents practical. However, as opposed to airlines, routing a shipment from New Englad to the Pacific Northwest via DFW, the railroads need to keep the operating milages as short as reasonably practical. Gateways other than Chicago also have their own set of capacity constraints.

Wall Streeet and it's mind set that the railroads are an archaic and dying form of transportation have directly caused the capacity constraints by pricing capital dearly. Railroads and not like the Dot.Com industry that can turn paper profits from minimal investment; railroads require real investment into real objects (Engines, Cars, Tracks, Signals, and employees to operate and maintain it all)....such investment in the Wall Stree World doesn't return itself 5 times over in the first year and is thus percieved tobe a poor investment and rated accordingly.

[Anonymous]

There's a number of issues here. I don't want to try to summarize the Chicago issue of Trains done two years ago, and if you're interested in Chicago, doing without that issue is like trying to build Boulder Dam without concrete. (Maybe if you had a 900-ton beaver with a 200-foot tail and a big pile of sticks ... )

One of the economic principles of railroading is "distance makes money, terminals cost money." A terminal cannot generate income, it is purely overhead. Thus a railroad spends the least possible money on terminals.

When the terminal also includes an interchange with another railroad, the costs and complexities grow by an order of magnitude. Now there are in effect two or more abutting terminals, each a gigantic money pit, that each have to coordinate the interests of two or more railroads, interests which will be congruent on through traffic and divergent on terminating and originating traffic. It's quite schizophrenic, and there's no way around it -- economics only apportions competing interests, it doesn't make them go away.

Because Chicago historically was more of a origin/destination for traffic than it was a waystation for traffic, there were very large economic disincentives for each railroad to spend money building efficient connections to the other railroads at Chicago. These connections weren't going to be used very much in relation to the use each railroad was going to get from its own main line leading to and from Chicago. So the historic pattern was to build enormous capacity right up to the city, and skimp when it got into the city, leaving somewhat of a moat between the railroads n the city itself. Whether one likes that or not, these companies would have been irrational to do otherwise. A company must spend most of its money on the things that generate most of the revenue.

Today, Chicago still is a huge consumer/originator of traffic. In terms of intermodal lifts each year, it is the number one city in the world, bigger than Hong Kong, Shanghai, Rotterdam, or Los Angeles. So a lot of the traffic going to Chicago is never going to come out the other side. That traffic creates zero incentive for bypasses or connections through the city. Moreover, that very huge weight of terminating traffic pulls a lot of the through traffic into Chicago with it. That's because traffic comes in trainloads, if it's to be moved at a profit, and trains that are short-tonnaged increase costs more than they increase revenues.

Suppose, for instance, Fresno, California, acting as a collecting point for a broad territory in the Central Valley, generates for BNSF enough carload traffic every day to fill a train for Chicago and eastern points, but about 50% of that traffic will stop somewhere in Chicago or its environs. Each Chicago car could conceivably end up on any one of the Class Is, belt lines, or short lines serving Chicago, or about a dozen different railroads -- and many of those railroads have several lines into Chicago. How do you handle this traffic? Your three simplest choices are:

1. Originate one train a day at Fresno, run it into Chicago, break it down there, then deal with getting the eastern carrier cars through Chicago to them. This is essentially the cheapest in terms of labor, but creates some serious delays in Chicago for the through cars. The terminating cars will do pretty well, though (back to favoring yourself first).
2. Originate one train a day at Fresno, run it to Galesburg, hump it, and when there are enough Chicago cars for yourself, run it, enough eastern carrier cars for a full train to NS, run that to NS connection, and ditto for CSX. But NS and CSX have more than one place they go out of Chicago, so Galesburg is going to have to classify the train into blocks for CSX or NS (each block requires its own bowl track). Conceivably this train could bypass Chicago, if and only if there happens to be some junction where CSX or NS can split the blocks apart without backtracking. Then CSX and NS are presented with the problem of sitting on that block until they get full tonnage, or combining it with other blocks from other railroads, or its own Chicago customers, to build a full train for each of its major destinations. That's beginning to sound like a yard, and if that yard is not in Chicago, it's either going to be sending cars backward to a junction in Chicago to get on the right line, or it will only be handling some of the blocks. So CSX or NS are going to have to classify this BNSF traffic at some point in Chicago, or just east of there. Conceivably, Galesburg could build, say, full trains for CSX-Selkirk, CSX-Queensgate, CSX-Evansville, etc.. Now, instead of Galesburg having one track in the bowl for CSX it empties out every few hours, or five tracks for CSX it empties out every 24 hours, it has just given over five or more tracks to CSX that won't empty out for several days. And instead of the cars dwelling in the yard for maybe 24 hours until there's a full trains' worth for CSX, the cars can easily dwell for 72, 96 hours, or more until enough accumulate for a train. (And customers like slow transit times how much?) You could run short-tonnage trains, but your crew costs go through the roof on both your road and on the connection, too, unless you're planning to simply substitute the interchange track for the bowl track, and after three or four days when a train's worth accumulates, CSX comes over and pulls it. If you hold the cars in Galesburg until a full train for a destination accumulates, you also start really increasing your equipment costs, too, because the cars aren't moving -- you just increased the number of cars you need to haul the same amount of freight. Empties will still be needed back in Fresno in the same number every morning.
3. Originate two trains daily at Fresno, one Chicago, and one east, and either double the crew costs to advance the cars at the same schedule -- chewing up twice the mainline capacity on which you can't run something else, or hold the cars in Fresno for 48 hours, average, instead of 24, which sucks up yard space in Fresno, runs up equipment costs, delays freight, etc.

What I'm trying to illustrate with this conundrum -- which is multiplied 10,000-fold by all the permutations of all the different main lines leading to Chicago -- is that there are no easy solutions. Any choice made has huge consequences. If Chicago consumed or generated no traffic, it would be less of a problem, it would look a lot like North Platte, Nebraska, except instead of having in effect five main lines entering from the west and three from the east, it would have 20 or so main lines converging. Dwell times would be pretty bad no matter what you did. However, because Chicago generates and consumes traffic, there's no way to get around running a lot of the trains into the city. And because each of the 20 or so main lines are sending traffic to each of the other 19 main lines, the logical place for a single yard -- to avoid backtracking -- is as close to the core of the city as you can get. If you don't put the yard there, then you start needing two, three, four yards, each one of them its own money pit, and the traffic flows instead of being ganged into one for maximum main line efficiency start getting fractured into several flows, requiring a lot more main track, crews, locomotives, etc.

There are very few traffic flows anywhere in the U.S. that go through Chicago (or could go around it) that are large enough to generate a through train every day, not without tolerating some enormous dwell times in yards. Only the unit coal trains that continue on to Detroit or Ohio or Northern Indiana are truly through trains, along with a very few intermodal and carload trains -- and most of those through carload trains are really Galesburg-Elkhart trains, or the equivalent, which means those cars have lots of dwell they're already carrying. It must be acknowledged that there are several hundred important origin-destination points and several thousand minor origin-destination points in the U.S. and Canada that send significant traffic to and through Chicago every day. Together, their business add up to 1,200 or so freight trains. While that sounds like a lot, individually by origin-destination pair, they each break down to flows of 5, 10, 20 cars a day. That many cars is not enough to build a train that can be operated at a profit.

Railroads make money by assembling many shipments into one collection called a train. Each of those individual shipments is yoked to every other individual shipment. If you want to run cars one at a time, and avoid all this problem with yards, connections, bypasses, terminals, interchanges, dwell time, etc., the transportation world already has a solution for you: it's called a truck. You'll gain time and certainty, and pay more in rate. If you want a low rate, you subordinate the independence of your single car to the interdependence of a train, and pay more in time, uncertainty, and unreliability. Unless you as a shipper can generate your traffic in unit-train quantities, you cannot expect the independence of movement of a truck at the low cost of a train.

OS

[oltmannd]

There is a whole $2B plan to detangle Chicago. The RRs and the city put it together. It benefits all the roads, Metra, Amtrak and the highways.

But, when you just barely have enough capital to keep things status quo, big projects like this dont' happen. CN, NS and BNSF might be able to scrape up some capital but it would help if CSX and UP could get straightened out. Having just one road opt out makes the whole thing impossible.

[Anonymous]

I vaguely remember an article in Trains about 30 years ago dealing with the interchange of cars between the then SF and the NYC. There was the Kankakee line which ran between Streator, IL. and South Bend, IN. which avoided going
into Chicago for interchange.

SF preferred Chicago since they would get more mileage on the shipment and therefore more money.

How short sighted. I believe Mr. Kneiling even agreed with SF.

Now, the Kankakee has portions torn up. How many trains that need to go Chicago today could possibly use this route which would be faster. I don't know. But it seems anything that can be diverted from Chicago would be an improvement.

Sadly, I feel the money won't come forward for what may be cheap solution and reduce at least of a little bit of Chicago's congestion.

[Anonymous]

QUOTE: Originally posted by O.S.


Would you like to know why railroad transportation is the way it is at Chicago and other gateway cities? If you do, I and others would be delighted to answer.

Yes I would like to know and I'm probably not the only one.

[Anonymous]

Sounds like it's time to build a new by pass around the windy city.

[Anonymous]

If that's what the Economist said, then the article is a screed, a diatribe, and arrogant. Or it got its information from a source with an axe to grind or with an undisclosed financial interest in a certain outcome.

What makes you think railroads aren't interested? No insult intended, but if you approach railroads and railroaders with the presumption that they're antiquated, dimwitted, or complacent, I don't think you'll have a positive experience. There are substantive economic reasons that created the way things are done at present. I recognize that street legend says railroads are the way they are because of (check one) venality, stupidity, arrogance, greed, or all of the above, but street legend is self-serving and uninformed.

Would you like to know why railroad transportation is the way it is at Chicago and other gateway cities? If you do, I and others would be delighted to answer.

OS

[Anonymous]

Last week’s Economist had an article on Chicago bogging down with 1,200 trains a day, a third of the nation’s rail traffic. It can take days to get freight across town. Unlike airlines, or any efficient logistics operation railroads have no intermediary to route shipments from one carrier to another. Railroads just fax news of arriving trains.

The Chicago Transportation Coordination Office (CTCO) wants to fix this and a planned Chicago Region Environmental and Transportation Efficiency (CREATE) program is waiting for federal funding (which they’ll never get from Bush).

I’d imagine anyone planning logistics at FedEx or WalMart would look at this and wonder why railroads have neglected hubs that connect carriers.

This looks like an obvious business opportunity for logistics entrepreneurs to save railroads and customers a huge amount of time and $ nationwide. Why aren’t railroads a lot more interested.